检索结果-内蒙古大学图书馆

4th International Conference on Information Science and Education, ICISE-IE 2023

作者： Zhu, Xiaoke Dong, Zhiwei Chen, Xiaopan Shi, Yufeng Henan Engineering Research Center of Intelligent Technology and Application Henan University Kaifeng China School of Computer and Information Engineering Henan University Kaifeng China Henan University Henan Key Laboratory of Big Data Analysis and Processing Kaifeng China

ISBN: (纸本)9798350394610

After the outbreak of the COVID-19 pandemic, online teaching has gradually emerged as an indispensable component of education. Despite its convenience, online education lacks the immediate interactive experience inherent in traditional classrooms. This paper presents a meticulously designed DS-YOLOv8 network model, specifically tailored for real-time recognition of students raising their hands under cameras. On our proprietary dataset, this network model exhibits exceptional accuracy in hand-raising detection, showcasing a 1.3% improvement in the AP50 index compared to the baseline YOLOv8 model. Our focus lies on the application of hand-raising detection within online classrooms with an objective to provide enhanced support for teachers by promptly identifying students raising their hands and thereby improving teaching efficiency while enriching the interactive experience of students in virtual learning environments. © 2023 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Federated Learning Based User Scheduling for Real-Time Multimedia Tasks in Edge Devices 1

引用

3rd EAI International Conference on Edge Computing and IoT, EAI ICECI 2022

作者： Wen, Wenkan Liu, Yiwen Gao, Yanxia Zhu, Zhirong Shi, Yuanquan Peng, Xiaoning School of Computer and Artificial Intelligence Huaihua University Huaihua418000 China Key Laboratory of Wuling-Mountain Health Big Data Intelligent Processing and Application in Hunan Province Universities Hunan Huaihua418000 China Key Laboratory of Intelligent Control Technology for Wuling-Mountain Ecological Agriculture in Hunan Province Hunan Huaihua418000 China

ISBN: (数字)9783031289903

ISBN: (纸本)9783031289897

Edge networks are highly volatile and the quality of device communication and computational resources change not only over time but also according to the movement of users. Current federation learning suffers from poor device network state and failure of devices to upload models in a timely manner. To address these problems, an intelligent scheduling mechanism that uses the predicted device state based on device information to select the appropriate device for federated learning is proposed in this paper. By focusing on information such as communication quality, computational resources, and location information, the information of edge devices is collected to analyze and predict the device network and computing resources to further analyze the state of devices in depth. Experiments are conducted on real datasets, and the experimental results show that the proposed scheduling method can make the global model fit faster than without the algorithm, which significantly improves the training efficiency of federated learning. © 2023, ICST Institute for computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

TMFIF:Transformer-based multi-focus image fusion 5

TMFIF:Transformer-based multi-focus image fusion

引用

5th International Conference on computer Vision, Image and Deep Learning, CVIDL 2024

作者： Li, Rui Geng, Shengling Zhang, Dan Zhou, Mingquan Qinghai Normal University School of Computer Science Xining China Nanyang Institute of Technology School of Computer and Software Nanyang China People's Government of Qinghai Province & Beijing Normal University Academy of Plateau Science and Sustainability Xining China Qinghai Normal University The State Key Laboratory of Tibetan Intelligent Information Processing and Application China

ISBN: (纸本)9798350373820

Multi-focus image fusion is a hot topic in the field of image processing, and it is a fundamental problem in the fields of image editing, image synthesis, and target retrieval. In previous fusion methods, although feature-rich datasets, models, and algorithms have been provided, there are still many problems with the effective fusion of distant and near views in complex backgrounds. To solve the challenging multi-focus image fusion problem more accurately, we introduce the Transformer Network based on Encoder Decoder (TMFIF), which can extract more generalized features based on the features of multi-focused images to achieve image fusion in complex backgrounds for better visual effects. In this work, we achieve fusion by inputting two images, near and far view. We compare the performance with other multifocal image fusion algorithms by conducting experiments on publicly available datasets and illustrated using existing evaluation methods and evaluation metrics;the results of the experiments show that our method visualizes the fusion effect better through the encoder and the decoder and the evaluation metrics are also relatively good. © 2024 IEEE.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

Online Q&A for Machine Learning Courses Based on Formula Recognition 4

Online Q&A for Machine Learning Courses Based on Formula Rec...

引用

4th International Conference on Information Science and Education, ICISE-IE 2023

作者： Zhu, Xiaoke Chen, Xiaopan Li, Minghao Zhang, Tianxiang School of Computer and Information Engineering Henan University Kaifeng China Henan Engineering Research Center of Intelligent Technology and Application Henan University Kaifeng China Henan University Kaifeng Henan Key Laboratory of Big Data Analysis and Processing China

ISBN: (纸本)9798350394610

Mathematical formulas are ubiquitous in courses such as Machine Learning. Understanding these formulas is the primary challenge for studying of students. If a searchable knowledge base can be built for these formulas, it will help students learn. Accurate identification of these formulas is an important step in building and querying the knowledge base. However, mathematical formulas have complex two-dimensional and nested structures, which makes them inconvenient to search like ordinary text. This paper employs Faster R-CNN to detect regions containing mathematical formulas and improves its backbone network, anchor box dimensions, and pre-training network to make it more accurate and efficient in identifing these formulas. Furthermore, in order to enhance the features and meaningful representation of mathematical formulas in the input image, we introduce the distance transformation. Finally, comparative experiments were conducted on the publicly dataset GTDB. Our proposed approach achieved recognition rates of 91.22% and 85.74% for isolated and hybrid formulas detection. © 2023 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

LSTM-based Pedestrian Trajectory Prediction Model under the General Direction Mechanism: DIR-LSTM 3

LSTM-based Pedestrian Trajectory Prediction Model under the ...

引用

3rd International Conference on Electronic Informationtechnology and Smart Agriculture, ICEITSA 2023

作者： Luo, Junyuan Fu, Qiongfang Xiao, Jianhua Xiao, Hongbo Liao, Jiyong School of Computer Science and Engineering Huaihua University Hunan Huaihua China Key Laboratory of Wuling-Mountain Health Big Data Intelligent Processing Application in Hunan Province Universities Hunan Huaihua China Key Laboratory of Intelligent Control Technology for Wuling-Mountain Ecological Agriculture in Hunan Province Hunan Huaihua418000 China

ISBN: (纸本)9798400716775

Accurate prediction of future traffic flow trends is essential to solve urban transportation problems. However, traffic flow prediction faces great challenges due to the multimodal nature of pedestrian behavior and the complexity of the traffic environment. Although a large number of studies have been conducted to investigate these issues in depth, there are still some limitations. In order to address these challenges more effectively, we propose a pedestrian trajectory prediction model based on long-short-term memory networks (LSTMs): the DIR-LSTM. The model introduces an innovative generalized direction mechanism and a self-attention mechanism, which captures pedestrian movement patterns more comprehensively and accurately by predicting overall directional movements first and then gradually subdividing them into individual directions of movement. The DIR-LSTM is designed to address the challenges posed by the diversity of pedestrian behaviors and the complexity of urban environments. To validate the state-of-the-art of the model, we conducted experiments using the publicly available ETH [10] and UCY [9] datasets. The experiments demonstrate that DIR-LSTM performs better in terms of accuracy compared to other models, providing a more reliable prediction tool for future urban traffic management. © 2023 ACM.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

TM2SP: A Transformer-based Multi-Level Spatiotemporal Feature Pyramid Network for Video Saliency Prediction

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年第6期35卷 5236-5250页

作者： Li, Chenming Liu, Shiguang Tianjin University School of Computer Science and Technology College of Intelligence and Computing Tianjin300350 China Tianjin University Tianjin Key Laboratory of Cognitive Computing and Application Tianjin300350 China

This paper proposes an end-to-end video saliency prediction network model, termed TM2SP-Net (Transformer-based Multi-level Spatiotemporal Feature Pyramid Network). Leveraging the strong encoding learning capability of Video Swin Transformer for video data, we design a Multi-level Spatiotemporal Feature Pyramid Network (MLSTFPN) that effectively detects and enriches salient regions and spatial details across different scales. In particular, a pre-trained image saliency detection encoder is employed to extract salient features from each frame, serving as prior knowledge to guide the multi-scale spatiotemporal feature fusion and decoding processes. Additionally, we introduce an Inception Gate-Controlled Fusion (IGCF) and Layered Self-Attention Aggregation Fusion (LSAF) mechanisms to efficiently merge spatiotemporal features across various stages. Finally, extensive experiments conducted on the DHF1K, Hollywood-2, UCF-Sports, and six audio-visual saliency datasets demonstrate the superiority of our method over existing state-of-the-art approaches. © 1991-2012 IEEE.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Edge Temporal Anti-Aliasing

Edge Temporal Anti-Aliasing

引用

2022 International Conference on computer Graphics, Artificial Intelligence, and Data Processing, ICCAID 2022

作者： Xu, Zhi Luo, Yuhang He, Daojing Qin, Yiping Wang, Xin Peng, SiTao Yao, JianFan School of Computer Science and Information Security Guilin University of Electronics Technology No.1 Jinji road Guilin541004 China Guangxi Key Laboratory of Precision Navigation Technology and Application Guilin541004 China School of Computer Science and Technology Harbin Institute of Technology Shenzhen518055 China China Nuclear Power Technology Researc Institute Co. Ltd. Shenzhen518028 China

ISBN: (纸本)9781510663350

Temporal Anti-Aliasing (TAA) is a popular method for eliminating temporal aliasing problems. However, the images simply processed by TAA become blurred and lose some details. In this paper, an improved TAA algorithm named Edge Temporal Anti-Aliasing (ETAA), is proposed. A time iterative edge detection method is designed to enhance the detection accuracy of pixels with temporal aliasing and spatial aliasing. These pixels are blurred and blended, and other pixels are replaced with the pixels of the current frame. Furthermore, an approximate minimum filter is used to eliminate the flickering phenomenon of high-energy noise. Compared to TAA, ETAA outperforms TAA in terms of image detail preservation when the rendering camera moves. Meanwhile, the time cost of proposed method also can satisfy the requirement of real-time rendering. Experiment results show that ETAA can effectively and quickly eliminate the time aliasing when the camera moves, and achieves a better trade-off in flicker, ghosting and blurring. © 2023 SPIE.

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

RDGFuzz: A directed greybox fuzzing optimization method based on Rich-Branch nodes 6

RDGFuzz: A directed greybox fuzzing optimization method base...

引用

6th International Conference on Electronic Information technology and computer Engineering, EITCE 2022

作者： Wu, Zejun Lu, Li Jia, Qiong Chen, Zhihao State Key Laboratory of Mathematical Engineering and advanced Computing Zheng zhou China National Local Joint Engineering Laboratory of Network Space Security Technology Zheng zhou China Beijing Institute of Computer Technology and Application BeiJing China

ISBN: (纸本)9781450397148

Directed fuzzing technology is one of the key technologies to quickly reach a specific location of software, and to conduct targeted testing or bug recurrence. However, directed fuzzing technology has some problems, such as unreasonable seed energy allocation, low code coverage and incomplete testing. To solve the above problems, this paper proposes an optimization method of directed fuzzing based on Rich-Branch nodes. In this method, the concept of Rich-Branch nodes is defined and the algorithm of extracting Rich-Branch nodes is given. The optimization method collects the coverage information of the target program in the running process, calculates the weights of covered functions and nodes in real time by combining CG and CFG of the target program, and generates a list of Rich-Branch nodes. According to the weights of Rich-Branch nodes, the seed energy allocation algorithm of AFLGo is optimized and improved. Compared with AFLGo, this optimization method improves the average code coverage of each targeted point by 56.79%, and has the same target reaching ability as AFLGo. © 2022 Association for Computing Machinery.

关键词： Software testing

来源：评论

学校读者我要写书评

暂无评论

SQAT-LD: SPeech Quality Assessment Transformer Utilizing Listener Dependent Modeling for Zero-Shot Out-of-Domain MOS Prediction

SQAT-LD: SPeech Quality Assessment Transformer Utilizing Lis...

引用

IEEE Workshop on Automatic Speech Recognition and Understanding

作者： Kailai Shen Diqun Yan Li Dong Ying Ren Xiaoxun Wu Jing Hu Faculty of Electrical Engineering and Computer Science Ningbo University Zhejiang Key Laboratory of Mobile Network Application Technology

In this paper, we propose the speech quality assessment transformer utilizing listener dependent modeling (SQAT-LD) mean opinion score (MOS) prediction system, which was submitted to the 2023 VoiceMOS Challenge. The system is based on a combination of self-supervised learning (SSL) models and listener-dependent modeling. Due to this challenge’s emphasis on real-world and challenging zero-shot out-of-domain MOS prediction in three different voice evaluation scenarios, we specifically designed a two-branch module to predict scores and weights for each frame, aiming to achieve better generalization. In the challenge, our system achieved fourth place in Track 1a, second place in Track 1b and first place in Track 2. Additionally, we conducted an ablation study to investigate the effectiveness of our proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Coupled Graph Convolution Network for Cross-Scene Multispectral Point Cloud Classification

Coupled Graph Convolution Network for Cross-Scene Multispect...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Mingye Wang Qingwang Wang Tao Shen Jian Song Kunming University of Science and Technology Kunming China Yunnan Key Laboratory of Computer Technologies Application Kunming China

Cross-scene multispectral point cloud classification aims to transfer knowledge of labeled source scenes to improve the discriminability of the model on the unlabeled target scenes. From a novel perspective, we argue that the information transfer between the source and target scenes can be used to solve cross-scene multispectral point cloud classification task. Specifically, we propose a Coupled Graph Convolutional Network (Coupled-GCN) to achieve joint alignment of node- and class-level structures within scenes by passing information between different scenes. To reduce the effect of spectral shift between the source and target scenes and seek scene-invariant intrinsic features, we propose a scene adaptive learning module by optimizing three different loss functions, namely, source classifier loss, domain classifier loss, and target classifier loss as a whole. In the cross-scene multispectral point cloud classification task, the proposed Coupled-GCN can alleviate the spectral shift problem compared to the traditional GCN and achieves an overall F_score of 65.04%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：